Why the same study pays more on one panel than another

The panel isn’t the client

Every time you take a paid survey, there’s a client somewhere behind it, usually a market research agency working for a brand that wants to know what people think about a product, an ad, or a pricing idea. That client doesn’t pay the panel you signed up on. It pays a fee per completed interview, and that fee gets sliced up before any of it reaches you.

The panel you’re logged into is one link in a supply chain: client, research agency, sample provider or exchange, panel, respondent. Each link takes a cut. How big that cut is, and how it’s calculated, is set independently by every company in the chain. That’s the entire reason the same underlying study can hand you a very different reward depending on where you access it.

How a study actually gets priced

A research agency has a budget for a project. Say they need 500 completed interviews from people who own a car and shop online at least weekly. They set a target cost per complete, then push that job out through one or more sample exchanges, companies like Cint, Lucid, or Dynata that connect research buyers to a network of panels. Panels plug into these exchanges through an API, sometimes called a router, and get offered a slice of the job.

Each panel negotiates its own margin with the exchange or the agency. A large, well-established panel with strong data quality scores can often negotiate a better split than a smaller or newer one, because buyers trust their traffic more and are willing to pay for it. A panel that’s desperate for volume might accept a thinner margin just to keep surveys flowing to its members. None of this is visible to you as a respondent. You just see a reward number on a tile.

This is the core mechanism behind survey panel pay differences. Each panel sits at a different point in the payment chain, holding a different negotiated cut of the same client budget. Generosity never enters into it.

The router problem

Because the same study can be distributed through multiple sample exchanges at once, it can genuinely appear on several unrelated panels simultaneously, sometimes with the same screening questions, sometimes with slightly reworded ones. If you’re a member of more than one panel, you may occasionally recognize a study you already completed elsewhere. The reward for it can differ by a wide margin between the two panels. Each panel cut its own deal with the exchange, and that deal sets your number. The questions are the same either way.

This also means a panel with a reputation for high engagement or fast fill rates can sometimes get priority access to better-paying jobs, while a panel that fills slowly or has quality flags gets offered the leftovers, or gets routed into studies after the higher bidders have already taken their share.

Screen-outs change the real math

Reward numbers are usually quoted for a completed survey. But most studies have quotas: they need a specific mix of ages, incomes, or behaviors, and once a quota fills, anyone who would have added to it gets screened out partway through, sometimes after five minutes of questions, sometimes after fifteen.

How a panel handles that screen-out matters enormously to your effective pay. Some panels pay a small flat amount for any screen-out past a certain point. Others pay nothing at all unless you complete. Some only pay for screen-outs on specific partner studies and not others, with no consistent policy across the board. A panel that quietly eats every screen-out as a zero is effectively paying you nothing for the time spent answering the qualifying questions, even if the headline reward for a completed study looks competitive.

This is one of the biggest hidden drivers of survey panel pay differences, and it’s almost never disclosed up front. You only find it by watching what actually lands in your account after a screen-out, not what the tile promised before you started.

Length of interview estimates aren’t always honest

Every study lists an estimated length of interview, usually in minutes. That estimate comes from the research agency’s own pilot testing, and it gets passed down the chain more or less unchanged. Some panels quietly pad it to account for their own respondents running slower. Others pass the agency’s number straight through, even when it’s optimistic.

If a study says eight minutes and consistently runs fourteen, your real per-minute rate on that panel is worse than the tile suggested, and it will keep being worse on every study from that same source. Panels that route more of their traffic through agencies with sloppy LOI estimates will systematically underpay relative to the time you actually spend, even when their advertised rewards look identical to a competitor’s.

Data quality scores affect what a panel gets paid

Sample exchanges score panels on data quality: how often their respondents get flagged for speeding through surveys, giving inconsistent answers, or failing attention checks. A panel with a poor quality score gets deprioritized for higher-value studies and may only get access to lower-paying, higher-volume work. This is part of why account health matters even if you never plan to abuse a panel. If enough other members on a panel behave carelessly and drag down its aggregate quality score, everyone on that panel can end up routed into worse-paying work, through no fault of their own.

Cash-out friction is part of the price

The reward you see isn’t the reward you keep. Minimum payout thresholds, currency conversion, gift-card-only redemption, and processing delays all reduce the effective value of what you earned. A panel that pays a slightly better headline rate but locks your balance behind a high minimum, or only lets you redeem in $25 gift card increments, can leave you worse off than a panel with a lower headline rate and instant PayPal cash-out. When comparing panels, the redemption terms are as much a part of the real pay rate as the reward number on the study tile.

How to measure the gap yourself

You don’t need to trust anyone’s claims about which panel pays better, including ours. The mechanism above tells you exactly what to track:

Time every survey from the qualifying question to either completion or screen-out, using a plain stopwatch or your phone’s clock.
Log the reward promised, the reward actually paid (including for screen-outs, if any), and the real time spent.
Divide paid amount by time spent to get your real rate per hour on that panel, not the advertised one.
Do this across a range of study types and lengths on each panel you use, since a single data point tells you nothing about a panel’s overall pattern.
Track how long it actually takes from hitting the payout threshold to money or a gift card landing in your hand.

Do this over enough surveys and you’ll see a real pattern specific to your own account, your own demographics, and the studies you happen to qualify for. Panel pay isn’t a fixed league table. It shifts by client budget, by the mix of studies running that month, and by your own profile, so the only measurement that means anything is the one you run yourself, on your own account, over time.

What this means practically

There’s no single panel that pays best across every study, every month, for every person. The gap between panels exists because of margin stacking, quota-driven screen-outs, LOI accuracy, and quality scoring, all of which are invisible from the outside and specific to the deal each panel has cut with its sample sources. Treat any reward number as a starting estimate, not a promise, and let your own stopwatch tell you which panels are worth your time this month.

If you want more of this kind of mechanism-level breakdown, of how panels actually work, how screen-outs get decided, and how to keep an account in good standing, Survey Farm covers it on the site and the channel.